ATAC-seq plasmid contamination after transfection

Dampor · 05-02-2017, 06:50 AM

HI all

I am trying to produce ATAC-seq libraries from transfected cells, but it seems a pain...

I have produced good libraries from non-transfected cells (sequencing was OK), but when I transfect cells with a plasmid I get, of course, a lot of tagmentation in the plasmid, such that most of my reads are plasmid-derived. I would like not to diminish the plasmid concentration, because all my experiment so far have been done using the same conditions.

We are trying to solve this problem but it does not look easy...

My idea is to get rid of the plasmid before tagmentation and probably use AMPure beads to suck it up just after lysing the cells. But I am not sure how this will work because there may be complications (I am going to test this next week, I think)

Another possibility is to check the tagmented genome (before amplification) on a gel and see if it is be possible to separate the plasmid from the genome. This actually raises the question if the genomic DNA is fragmented after tagmentation (I think it is, unfortunately... can you confirm?)

We were also considering sequence capture to get rid of the plasmid fragments, but this also has some problem in the fact that the adapters may then hybridise together resulting in non specific sequence capture.

Please let me know if you have found such a problem and came up with a solution.
Very much appreciated

Bests
Dam

rbd · 05-03-2017, 09:47 AM

This isn't something that I have tried but I have thought about it. If your plasmid isn't too long you could possibly try using long biotinylated oligos that will hybridize to known sequence in your plasmid. You then pull down plasmid DNA fragments from your ATACseq sample post-tagmentation, pre-amplification.

I would be interested to hear other people's thoughts about the viability of such a method.

Also there is another method which is likely to be more viable since it has already been published. I think there are some papers out using CRISPR/Cas9 and sgRNAs to deplete mitochondrial DNA in ATACseq samples. I don't see why that can't be applied to deplete your plasmid DNA.

check out this link

http://biorxiv.org/content/early/2016/11/22/087890

Also, to answer your question, the DNA is fragmented after tagmentation.

nucacidhunter · 05-04-2017, 05:02 AM

It might be possible to adapt NuGEN AnyDeplete technology to get rid of plasmid sequences.

Sequence capture will not work as only one strand of plasmid read will be captured and the other strand will stay in supernatant along the host fragments which will be amplified or sequenced. If you use host probes for capture it might be expensive as probes have to be synthesised for whole host genome.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
RNAseq normalize DE by edgeR with transfection positive control gene	jsage8	Bioinformatics	0	06-08-2015 11:25 AM
Plasmid contamination in Long Tag Paired End library	kmcarr	454 Pyrosequencing	3	03-11-2009 05:29 PM

05-02-2017, 06:50 AM	#1
Dampor Member Location: UK Join Date: Jun 2012 Posts: 14	ATAC-seq plasmid contamination after transfection HI all I am trying to produce ATAC-seq libraries from transfected cells, but it seems a pain... I have produced good libraries from non-transfected cells (sequencing was OK), but when I transfect cells with a plasmid I get, of course, a lot of tagmentation in the plasmid, such that most of my reads are plasmid-derived. I would like not to diminish the plasmid concentration, because all my experiment so far have been done using the same conditions. We are trying to solve this problem but it does not look easy... My idea is to get rid of the plasmid before tagmentation and probably use AMPure beads to suck it up just after lysing the cells. But I am not sure how this will work because there may be complications (I am going to test this next week, I think) Another possibility is to check the tagmented genome (before amplification) on a gel and see if it is be possible to separate the plasmid from the genome. This actually raises the question if the genomic DNA is fragmented after tagmentation (I think it is, unfortunately... can you confirm?) We were also considering sequence capture to get rid of the plasmid fragments, but this also has some problem in the fact that the adapters may then hybridise together resulting in non specific sequence capture. Please let me know if you have found such a problem and came up with a solution. Very much appreciated Bests Dam

05-03-2017, 09:47 AM	#2
rbd Junior Member Location: Concord, California Join Date: Oct 2014 Posts: 6	This isn't something that I have tried but I have thought about it. If your plasmid isn't too long you could possibly try using long biotinylated oligos that will hybridize to known sequence in your plasmid. You then pull down plasmid DNA fragments from your ATACseq sample post-tagmentation, pre-amplification. I would be interested to hear other people's thoughts about the viability of such a method. Also there is another method which is likely to be more viable since it has already been published. I think there are some papers out using CRISPR/Cas9 and sgRNAs to deplete mitochondrial DNA in ATACseq samples. I don't see why that can't be applied to deplete your plasmid DNA. check out this link http://biorxiv.org/content/early/2016/11/22/087890 Also, to answer your question, the DNA is fragmented after tagmentation.

05-04-2017, 05:02 AM	#3
nucacidhunter Jafar Jabbari Location: Melbourne Join Date: Jan 2013 Posts: 1,238	It might be possible to adapt NuGEN AnyDeplete technology to get rid of plasmid sequences. Sequence capture will not work as only one strand of plasmid read will be captured and the other strand will stay in supernatant along the host fragments which will be amplified or sequenced. If you use host probes for capture it might be expensive as probes have to be synthesised for whole host genome.